Aniseikonia testing in an adult population using a new computerized test, "the Aniseikonia Inspector"

To determine the measurement characteristics of a new computerized test, the Aniseikonia Inspector Version 1, on a sample of clinic patients.

Aniseikonia was measured in the vertical, horizontal, and oblique meridians on 320 patients (mean 55 years old, range 17-89 years) prior to their optometric exam using the psychometric methods programmed into the Aniseikonia Inspector Version 1. Statistical analyses were performed to determine the distribution of aniseikonia in the sample of patients and the relationships between the amount of aniseikonia and patients' habitual refractive correction, visual acuity, stereopsis and binocular alignment status. The characteristics of the individual measurements were also examined.

The means and standard deviations of the measured aniseikonia in the vertical, horizontal, and oblique meridians were -0.5% (2.5%), -0.1% (3.3%) and 0.3% (2.8%) respectively. The means in the vertical and oblique meridians were significantly different from 0.0 (p=0.0001, p=0.0314) while that in the horizontal was not (p=0.61). The distributions of aniseikonia showed that 65.6%, 57.5% and 64.3% had within +/-1.0% aniseikonia in the vertical, horizontal and oblique meridians, respectively. Correspondingly, 16.9%, 25.6% and 25.8% had aniseikonia of +/-3.0% or greater. The discrepancy between these percentages and those expected in a normal distribution indicate that the distributions were significantly more peaked than a normal distribution. This departure from normal is due to a few extreme values in the tails. The magnitude of aniseikonia had no statistically significant relationship with the patients' habitual refractive correction, visual acuity or stereopsis. The effect of phoria on the amount of aniseikonia was significant, more so for measurements in the horizontal meridian. The individual measurements, which are the average of two trials using the method of adjustment, showed no significant bias, no relationship between the means and differences in the two readings, but large differences between the two readings. Measurements in the vertical direction seem to be more stable than those in the other two meridians.

As measured with the Aniseikonia Inspector 1.0, the majority of the patients sampled in this study exhibited 1.0% or less aniseikonia and were therefore not likely to have symptoms related to aniseikonia. At least 17% of patients had 3.0% or greater aniseikonia measured in the vertical meridian. The Aniseikonia Inspector warrants further evaluation in a clinical setting because of the large limits of agreement between the two settings that are average to determine the magnitude of the aniseikonia. These limits differ considerably from those established by the designers and, therefore, raise questions regarding the actual resolution of the instrument as compared to the nominal resolution.